Probabilistic data association filter

The probabilistic data association filter is a statistical approach to the problem of Radar tracker#Plot to [track association|plot association] in a target tracking algorithm. Rather than choosing the most likely assignment of measurements to a target, the PDAF takes an expected value, which is the minimum mean square error estimate. The PDAF on its own does not confirm nor terminate tracks.
Whereas the PDAF is designed to track only one target in the presence of false alarms and missed detections, the Joint Probabilistic Data Association Filter can handle multiple targets. The first real-world application of the PDAF was probably in the Jindalee Operational Radar Network, which is an Australian over-the-horizon radar network.

Implementations

MATLAB: The PDAF and JPDAF algorithms are implemented in the singleScanUpdate function that is part of the United States Naval Research Laboratory's free Tracker Component Library.
Python: The PDAF and other data association methods are implemented in Stone-Soup. A tutorial demonstrates how the algorithms can be used.